Polarforschung71 (3), 79 - 95,2001 (erschienen 2003)
Reeonstruetion of Outlet Glaeier Tongues of the lee Age South- Tibetan lee Cover
between Cho Oyu and Shisha Pangma
as a further Proof of the Tibetan Inland lee Sheet
by Matthias Kuhle'
Abstract: The author has been engaged in the rcconstruction of the !ce Age glaciation and snow line depression carrying out over 20 expeditions and re- search trips to Tibet and its surrounding mountains since 1976. From the field observations made in 1996, two out let glacicrs were reconstructed: the ca,75 km long Bo Chu- and the ca. 100 km long Kyetrak Chu glacier. Their sources were not in the Himalayas, but further north ofits main ridge in S Tibet. From this location they flowed down through the Himalayas to the south slope. Their past existence and run-off ovcr the local water dividc in S Tibet (for the Bo Chu glacier) and the Himalayas (for the Kyetrak Chu glacier) provide evidence of important ice masses on the Tibetan plateau(er Fig. I complex I3 between Shisha Pangma and Mt. Everest).
Zusammenfassung: Seit 1976 hat der Verfasser zur Rekonstruktion der eiszeitlichen Vergletscherung und Schneegrenzdepression Über 20 Expedi- tionen und Forschungsreisen nach Tibet und in seine Randgebirge durchge- führt. Die hier vorgelegten Feldarbeitsbefunde von 1996 rekonstruieren zwei Auslassgletscher. den ca. 75 km langen Bo Chu- und den ca. 100 km langen Kyetrak-Gletscher, welche nicht vom Himalaya, sondern nördlich seines Hauptkammes von SÜdtibet ausgegangen und durch den Himalaya hindurch bis in dessen SÜdabdachung hinabgeflossen sind. Ihre vorzeitliche Existenz und ihr Abfluss Über die lokale Wasserscheide in SÜdtibet - was den Bo Chu- Gletscher betrifft - und Über die des Himalaya - was den Kyetrak-Gletscher betrifft - liefern den Beweis für bedeutende Eismassen auf dem Tibetplateau (vgl. Abb, I Inlandeiskomplex I3 zwischen Shisha Pangma und Mt. Everest),
PROBLEM
The author has been working for the last 27 years on the question of the extent of glaciation of the Himalaya, Kara- korum, Kuenlun, Quilian Shan and Tibet during the Pleisto- cene lee Age which is gradually being answered by information gained from Quaternary-geological und geomor- phological key locations. A suitable area for investigation is the south margin ofTibet where the upland lies adjacent to the Himalayas which are 3000 m higher (Fig. I, 3). Due to the southerly position at 28"N, the snow line (ELA) runs the highest here - nowadays and probably also during the LGM.
Of all the plateau areas, southern Tibet is therefore furthest away climatically from glaciation now and in past times, If even southern Tibet was covered by ice, then this must have been true also for central and northern Tibet because, for planetary reasons, the snow line dips towards the north. The idea is feasible because the Central Plateau which lies north of south Tibet is at the same height above sea level 01'high er.Itis therefore hypsometrically closer to past glaciation.
Geographisches Institut der Universität, Goldschmidtstr. 5, D-37077 Göttingen, Gerrnany: <mkuhledi.gwdg.de>
Manuscript received 26 March 2001, accepted 12 Mareh 2003
Present-day glaciers flow down from the highest summits of the Himalayas in all directions. On the north slopes, their tongue ends reach down to the level of the plateaus and into the high valleys of south Tibet, such as Yepokangara glacier on Shisha Pangma, Rongbuk glacier on Mt. Everest and Kyetrak glacier on the NW-side ofthe Cho Oyu (Fig. 12 below No. 1).
The large transverse valleys Iying between such mountain glacier areas which run down from Tibet and divide the Hima- layas into separate massifs, are not glaciated at present (e.g.
Tamur valley, Arun valley, Bote Chu, Marsyandi Khola, Thak Khola, Bheri Khola and Alaknanda valley). An important question is: Were these valleys glaciated during the lee Ages?
In contrast to the weak traces of glaciation in high Iying areas which are difficult to prove, the work of past valley glaciers can clearly be recognized. This difference is due to the follow- ing factors: Upland ice is cold based glacier ice, because the level of the upland runs about01'even above the snow line. As in the Antarctic 01'parts of Tibet, it has overlain permafrost.
Cold based ice usually freezes permanently to the ground. H, because of the shearing strain, it sometimes suddenly moves, it does not leave a smooth surface like warmer glacial ice resting on glacier meltwater but rather roughnesses which cannot be distinguished from weathering traces left by frost.
In the valleys, in contrast to the uplands, the flow velocity of the ice is higher due to reduction of the cross-profile of the outlet. Furtherrnore, the valleys lead down to lower areas which are far below the snow line and permafrost limit. Their glacier filling therefore consisted ofwarm based ice. Such fast flowing glaciers running over a film of meltwater leave the rocks round and smooth with polishing. Evidence of this are features such as roches moutonnees, polished rocks and glacier striations.
In contrast to the upland ice, which is found at the altitude of the snow line and in the nourishment area and therefore has mainly an erosive effect, the lower parts of the valley floors and flanks are covered by ground moraines. Their increasing thickness can be explained by the progressively positive mass balance of moraine material below the snow line in the direc- tion of the lowest ice margins. Such ground moraine covers preserve the polished forms of the rocks. The polishings are only visible and can only be proved beyond doubt on those rocks which were just recently relieved of the overlying moraine,i.e,have not yet been destroyed by weathering.
HIGH ASIA
Exaggeration 1:15 POintof View 350000 m a.8.1.K2 (Chogori.861601)
Nang~l~~~~at Maximum extension of the Pleisto-
11, 12, 13 ce ne inland glaciation 01 High Asia 11
Tsaidam Becken
I
(2620-300001)
Ouilian Shan(5808m}
Kakitu(570401) Animachin (628301)
"~, I I I r~~ I
KolaKang~~!
Anna- Shisha Namcha (775401)
purna Ipangma Bawa
l8091m) (804601) (775401)
I
' I
Mt. EverestManaslu (887401)
f8128m} Himalaya
Dhaulaqiri(8172m)
I
13 12
Nanda Devi (7816m) Tian-Shan Pamir
I
Pik PobedyMuztag Fang (7439m) (755501)
Pik Lenin
I
(7134m)
j
Pik Kommunismus
(7495m) .
Dra'it- M. Kuhle
Fig.l: The high glacial Tibetan ice had an extension of more than 2.4 million km'. The three centres ofglaciation 11, 12 and 13 were separated from each otherby the Tsaidam lake and the Tsangpo valley.
Abb. 1: Das Über 2,4 Millionen km' ausgedehnte hoch eiszeitliche tibetische Inlandeis hatte drei kuppelförmige Zentren Il,I2 und 13. Diese waren voneinander durch den Tsaidam-See und das Tsangpo-Tal getrennt.
8000 E 6000 zs:r4000
2000 00
Exaggeration1:20 A1tun Shan
Profile Muztag Feng - Mayer Kangri • Mt. Everest Profil Muztag Feng - Mayer Kangri - Mt. Everest
sec
Distancein km
Transhimalaya
1000
Tibet Himalaza
3000 Cartography: A FlemnilZ Oraft M.Kuhle
Fig. 3: Cross-profile ofthe Pleistocene inland glaciation ofTibet according to KUHLE (1998). This older reconstruction is supported in this paper by the observa- tions regarding the southern margin of Tibet, which are evidence ofthe transfluences ofthe outlet glaciers through the Kyetrak Chu (valley) and the Bo Chu (val- ley) west ofMt. Everest over the watershed ofthe Himalaya and into its southern slope.
Abb. 3: Querprofil des pleistozänen Inlandeises in Tibet nach KUHLE (1998). Diese ältere Rekonstruktion wird durch die in dieser Arbeit vorgestellten Daten vom SÜdrand Tibetsgestützt,Die Untersuchungen beweisen die Transfluenz von Auslassgletschern, welche dem Verlauf des Kyetrak Chu (Tal) und des Bo Chu (Tal) westlich des Mt. Everest Über die Wasserscheide des Himalayas bzw. in seine südliche Abdachung folgten.
As for the empirical proof of a former ice cover the search for traces of a past glaciation in these transverse valleys is highly promising. For this reason it must be established whether local glacier ice from the Himalayan mountains or far-travelled ice from the Tibetan plateau flowed through them. In the latter case, these would have been typical outlet glaciers which arose from a Tibetan inland ice cover at the southern edge of the plateau and then flowed down through the Himalayan trans- verse valleys. To provide evidence of the existence of such an outlet glacier, the most interesting factor would not be the
former glaciation of the valley but rather the glaciation at its source somewhat north of the Himalayas. Only a glaciation of the valley head would confirm Tibet to be the glacier catchment area and not the higher Himalayas which continue down-valley. The plateau region of Tibet with an altitude of about 5000 m would, in this case, have been above the Ice Age ELA - naturally not as high above it as the summits of the Himalayas but instead extended over a wide area. A plateau that rises above the snow line has the type of relief which can accumulate the most snow to feed a glacier. Since it does not
slope or, at most, only slightly, the ice flow is held back until a thickness of at least a few hundred metres is attained. The secondary increase in height due to such a plateau ice furthers the feeding of the glacier by lowering the annual surface temperature. This is accompanied by an increase in the annual snowfall. Thus, owing to its much larger area, the feeding by such a plateau glaciation exceeds that of 2000-3000mhigher summits by far.
The idea that from the southern edge of the Tibetan plateau antecedent transverse valleys have come down through the Himalayas is highly simplified. Strictly speaking, the valleys start from local water divides between the Himalayan south side and the Tibetan drainage system. Because the internal drainage in south Tibet takes place along shallow valleys, we are not talking about a plateau area in a strictly geometrical sense. Actually, it is a slightly dissected upland that might have been covered by an ice sheet. If so, the past outlet glaciers must have flowed down over transfluence passes.
In order to prove the existence of glaciers which arose from the ice of a Tibetan plateau which in its turn was also the prerequisite for the nourishing of the glaciers, two examples have been chosen.
CHOICE OF TEST VALLEYS ON THE MARGIN OF TIBET AND APPROACH TO INVESTIGATIONS
Two valleys between Shisha Pangma and Mt. Everest (Fig. I) were chosen, the axes ofwhich cross the Himalayas from Tibet in a southerly direction. One is the Bo Chu (Bote Chu; Pa Ho on the ONC map I: 1,000,000, H9, 1978) which runs across the main ridge of the Himalayas between Shisha Pangma und Chomolung Kang from Yagru Xiong La (Fig. 2, No. 25) to Dram (Zhangmu) (Fig. 2, No. I). The second valley is the Kyetrak Chu between Chomolung Kang and Cho Oyu which rises from the settlementofTing-Jih(Fig. 2, above No. 39) in a south direction to N angpa La (No. 3 I). Its axis continues south of this glacier pass in Nangpo Dzangpo to the settlement of Thame on the Hirnalayasouth-slope. Ifthe northerly extension of the valley axis of the Bo Chu over the Yagru Xiong La (pass) into the Xaga Chu is considerecl, the courses of the two valleys show fundamental similarities. Both of them cross local passes and water divides on their way out of south Tibet.
Here, the special topographical conditions mentioned in Section I are realized: the valleys lead out from the dissected Tibetan upland right across the Himalayas and slope steeply down to the lowland (cf density of the topographic contour line and altitudes of the top section of Figure 2 with those of the bottom section). One difference between them (which, however,can be neglected in this context) is that the Kyetrak valley is still partly glaciated whereas the Bo Chu is not. The present-day Kyetrak glacier flows from the Himalayas (Cho Oyu 8201 m) for 10 km in a northern direction into Tibet (Fig.
2, No. 31-32).
In accordance with the problem presented in Section I, the following questions were to be answered: (i) Was Bo Chu glaciated, i.e. what indications prove01'disprove a past glacia- tion of the valley?(ii) If Ice Age glaciation can be evidencecl, did its glacier come down from Tibet or was it only local ice from the still glaciated Himalayan mountains? (iii) Are there
any indications that the Kyetrak glacier might not have existed in its present form, thickness and flow direction? Could there possibly be evidence of a much more important ice stream of the type "outlet glacier"which might have filled the Kyetrak valley from Tibet? If so, such an ice accumulation would have flowed in the opposite direction to the present-day Kyetrak glacier over the present ice divide of the Himalayas (over the Nangpa La: Fig. 2, No. 31) in a southerly direction.
EVIDENCE OF A PAST GLACIATION OF BO CHU (ALSO BOTE CHU; PA HO OR SUN KOSI KHOLA), PROBABLY FROM THE LAST ICE AGE (LGM)
Indications ofglaciation in the lower Bo Chu
In the valley section of Lamosango (27°40-48'N, 85°45-55'E), the talweg ofthe Bo Chu runs at 700-900m a.s.l.. Here, 3-4 m long, erratic augen-gneiss boulders from the Shisha Pangma (for the petrography of the Shisha Pangma cf KUHLE 1988:
483, Fig. 43) were found. They lie on outcropping schists and metamorphicsilt-stones. Many ofthese boulders are rounded.
Others show subglacially formed potholes. Some of them can be found in the matrixof the morainic fine material. A further indication that the Bo Chu glacier tongue reached all the way down to the valley chamber is the occurrence of Pleistocene lateritic red weathering which sets in abruptly and over large parts, down-valley. Up-valley from the settlement of Barabise, rochemoutonnee-like glacier polishings are preserved up to at least 250 m above the talweg. They are proof of corresponding minimum thicknesses of ice in the valley. In the slope-depres- sions Iying between them, glacigenic layers of boulder clay metres- to decarnetres-thick, next to covers of autochthonous slope debris have been preserved. Based on this evidence, it is certain that at the time of its greatest length, the Bo Chu glacier ended in this valley chamber between 700 and 900 m a.s.l..
In the next valley section which stretches in an upward direc- tion to the junction with the Chaku Khola, glacigenic flank polishing occurs on both slopes. This is proved by the clear roundings of the outcropping edges of the strata. The follow- ing valley section upwards in direction of the settlement of Jhirpu presents two closely interlocking valley cross-profiles, The higher one shows the broad U-profile of the glacier bed, whereas the other one is set into the valley bottom in the form of a narrow V-profile due to subglacial erosion from the melting ice. The latter is typical of Ice Age glacier tongues which have flowed downwards as much as 3000 m below the ELA. Therefore, this V-profile has also been observed in other Himalayan transverse valleys (KUHLE 1982). Further up, the valley crosses the main Himalayan ridge so that the slopes are much higher and longer than they are in the lowland of the Himalayas. This causes a more intensive reshaping of the past glacigenic flank polishings by present-day flushing-out. For this reason, postglacial breakages of the rock interrupt the glacigenic polishings in some places. In this part of the Bo Chu, moraines are only to be seen as remnants. They have been eroded by monsoon-induced torrents. In many places, their substrate has been dislocated by linear mudflows and finally completely removed at the bottom of the valley by the Sun Khosi river.
Further evidence of an important lee Age valley glaciation is given by the glacially truncated spurs between the junctions with the side valleys. Triangular-shaped slopes such as these form parts ofthe flanks ofthe Bo Chu (Fig. 2, No. 4).
Glacigenic flank polishings occur also on both sides of thc Bo Chu between Lartza and Dram (also Khasa or Zhangmu). On the orographie right side opposite from Dram there isapartic- ularly important indication of a valley glaciation during the lee Age (Fig. 2, No. I). Itis an extensive glacial flank polish- ing up to 400 m above the talweg which is weil developed in the resistant metamorphie rocks of the Khumbu and Kath- mandu covers (KU 1-3; KN 2, 3 according to HAGEN 1969:
129). Itproves that the surface of the Bo Chu glacier rnust, at one time, have been at 2400 m a.s.l.. The smoothly polished rock surfaces, darkly striped from the influence of water, have been roughened by late-glacial and Holocene to present-day crumblings (Fig. 2, No. 2). Spallings cause small overhangs under which the rock has not been darkened by water. They can therefore be seen very clearly as large, light roughnesses in the rock, several square metres in size.
The transverse valley profile in the Bo Chu between Dram and the junction with the Fuqu Chu also provides evidence ofa past valley glaciation. In the middle of its course it IS U- shaped (at 28°02'N, 85°59'E, 2700 m a.s.l.; Fig. 2, No. 3). In some seetions it shows the characteristics of a gorge-like trough, i.e. ofa V-shaped valley, the flanks of which have a slightly concave form caused by abrasion ofavalley glacier (for details ofthis valley type cf.KUHLE 1982). The neighbour- ing side valleys (Fig. 2, No. 5 and a further side valley to the right of No. 3) also show both of these types of glacial valley cross-profiles.
One such mudflow (July 1996) brought down morainic mate- rial originating from the westerly tributary valley, Jangbo Khola. The mudflow completely destroyed the settlement of Lartza at 1300m a.s.l. at the bottom ofthe Bo Chu (45 people were killed). Far-travelled granite and augen-gneiss boulders, 3 x 4 x 5 m in size, were incorporated into the mudflow which derived from the massifs of the Shisha Pangma and Rolwaling Himal, that is from the upper catchment area of the Bo Chu.
The clayey-Ioamy matrix of the mudflow is typical mo raine matrix which was also taken up and displaced.
The largest moraine boulder transported by mudflow in this valley seetion ofthe Bo Chu is 6.7 x 11.3 x 17.7 m in size. It lies 6 km downstream from the settlement of Kodari, a few metres above the talweg. Xu DAOMING (1988) also describes this boulder as a displaced moraine component. According to his investigations, the boulder was transported by two mudflows (from the region depicted in Fig. 2, No. 3-5) caused by outbursts of moraine lakes in the Rolwaling massif in 1964 and 1981.
At 3680 m a classic "riegel" mountain has been polished by the glacier and shaped into a roche rnoutonnee (Fig. 2, No. 6).
At its culmination there are erratic augen-gneiss boulders (see above). At the junction with the Fuqu Chu end moraines have been deposited from the two Shisha Pagma south glaciers, presently separated (Fig. 2, No. 7). Because of their brownish weathered surfaces and their calculated ELA-depression compared with the present snow line of ca. 600-800m, they o
(j
.'
.']
)
c;
.2
.0~
'"
-g
'"
'"
:cc
U)
'&
u
must belong to the last Late-glacial period, i.e. they are sorne- what older than 13000-14250 yr BP (as for the chronological classification of the ELA-depressions see KUHLE 1997 Tab. I stage III-IV).
The Ice Age glaciation of the lower Bo Chu dealt with so far could possibly have arisen exclusively from the Fuqu Chu.
One argument in favour of this would be its connection to the very highest glacial catchment area of the Bo Chu, i.e. the 8046 m high Shisha Pangma (cf. Fig. 2). This is, however, not the case, because the middle part of Bo Chu was also filled with ice.
CUMULATIVE FREQUENCY GRAIN-SIZE CURVE 21.08.199611 Clay
< 2 2 - 6 6 - 20 20 - 60 60 - 200 - 600- 200 600 2000 (DIAMETER 1/1.000)
HUMUS cONTENT: 0,77 % LlME CONTE NT: 0,18%
Fig. 5: Morphometrie quartz grain analysis of 10 representative sampIes from south Tibet (cf. Tab. land Fig. 2, 4, 6-9).
Abb. 5: Morphoskopisehe Quarzkorn-Analyse von 10 repräsentativen Proben aus SÜdtibet (vgl. Tab. 1 und Abb. 2,4, 6-9).
Ddull [aeolian)/
lustrous (fluvial)
% 100
90 80 70 60 50 40 30 20 10 0
~ :0
s
~ ~0- 0- 0- 0- 0-
Cl) Cl) CD co Cl)
0 q 0 0 0
L!) r-, co 0:- e)
0J 0J C'J 0J C0
l1llilglacially crushed/
freshly weathered
% 100
90 80 70 60 50 40 30 20 10 0
~ 0J 0J
X :0- :0- :0-
0- 0- 0- 0-
Cl) Cl) Cl) Cl)
0 q 0 0
C0 C0
0J 0J 0J 0J
Fig. 4: Sediment samplc taken from a depth ofO.15 m at 3835 m a.s.l. on the orographie right side of the Bo Chu near the monastery of Milaripa. Loeality:
Fig. 2, No 11; moraine matrix of a high to late glaeial ground moraine to late- ral moraine terraee which has also bcen reworked glaciofluvially (cf. Tab I, Fig. 5, 21.08.96/1). A large part of the moraine is built up by erratie augen- gneiss substrate, resulting in a coarse-graincd matrix and a low peak of the fine grain in the c1ay.
Abb. 4: Sedimentprobe in 3835 m Ü.M. orographisch rechts im Bo Chu nahe dem Kloster von Milaripa aus 0,15 m Tiefe entnommen. Lokalität: Fig.2, NI'.
11; Moränenmatrix aus einer hoch- bis spätglazialen Grund- bis Ufermorä- nenterrasse, welche auch glazifluvial Überarbeitet worden ist (vgl. Tab. 1, Abb.
5,21.08.96/1). Die Moräne ist großteils aus erratischem Augengneisssubstrat aufgebaut, weshalb die Zwischenmasse grobkörnig und der Feinkorn-Peak im Ton niedrig ausgefallen sind.
The fact that the reshaping of the Last lee Age ground moraines in this middle part of the Bo Chu was caused by meltwater from late-glacial up to neoglacial glaciers can be proved by relatively young end moraines in the side valleys, because these are evidence of the typical late-glacial to neoglacial ice margins. Two ofthe end moraines come down to 4100 m a.s.1. in an orographie left-hand tributary valley (Fig.
From the Fuqu Chu junction upwards in direction of the Bo Chu, between 3670 and 3800 m, another well-preserved example of glacial polishing by the main valley glacier (No.
10) was found. On the main valley floor between the settle- ments of Nylamu and Kum Thang, a ground moraine from the maximum glaciation (LGM) (3700-4120 m a.s.l., 28°15' 20"N, 86°00'30"E; Fig. 2, No. 11) has been observed. The moraine contains isolated, far-travelled granite- and augen- gneiss boulders, sometimes up to one metre in length. These rocks occur as bedrock in south Tibet (KUHLE 1988, Fig. 43).
The moraine was first washed out superficially by meltwater ofthe late-glacial glacier. Later it was reshaped glaciofluvially into today's terraces. The clay peak and bimodal grain-size distribution (Fig. 4) documents the moraine character of the fine material matrix. The conspicuously large proportion of middle sand (41 %) on the moraine surface is evidence of fluvial reshaping. A proportion of 0.18 % calcium carbonate is proof of incorporation of only a small amount ofbedrock from the underground. The 200 SiO, grains analysed microscopi- cally (Tab. 1, Fig. 5,21.8.96/1) show a predominance of 62.5
% of the group glacially crushed/freshly weathered, typical of ground moraine. The 37.5 % quartz grains included in the morphoscopic group dull (aeolian) /Iustrous (fluvial) are proof of the late-glacial glaciofluvial and Holocene cold-arid reshap- ing of this ground moraine terrace. The flanks of this valley seetion have been polished back to a trough. The widening of the valley cross-profile was glacigenic as can be recognized from the glacially truncated spurs between the orographie left- hand junctions of the tributary valleys (Fig. 2, No. 12) which are covered with remnants of ground moraine.
The following discussion deals with the 23 km long seetion of the main valley between the two large orographie right-hand side valley junctions. These valleys join the Bo Chu from the Shisha Pangma massif. The southern valley is the Fuqu Chu (Fig. 2, No. 7), whilst the northern has no name (No. 20).
Indications ofa past glaciation in the middle Bo Chu
Directly opposite the junction with the Fuqu Chu, the smooth- ly polished surfaces of the orographie left-hand main valley slope (Bo Chu) are covered by ground moraine several metres thick, containing erratic boulders (Fig. 2, No. 8). Wherever this cover of lodgement till is missing, glacial polishings on relatively rapidly weathering rock faces are preserved. This is evidence of quite a young main valley glacier. A rather impor- tant ice thickness of at least 500 m is proved by the glacigenic rounding of the valley flank up to its culmination. At the top there is a glacigenic transfluence pass (Fig. 2, No. 9).
Sample No.! date 0,2 - 0,6 mm counted glacially crushed/ dull (aeolian) lustrous (fluvially remarks Probennr./ Datum quartz grains freshly weathered (in situ) äolisch mattiert polished) Anmerkungen
0,2 - 0,6 mm ausgezählte glazigen-gebrochen/ fluvial poliert Quarzkörner frisch (in situ) verwittert
XI 201 49.6% 26.8 % 23.6 % fluvial reworking more distinct than aeolian - Fluviale
Überarbeitung ausgeprägter als äolische
21.08.96/ 1 200 62.5 % 10.0 % 27.5 % 90 % quartz portions (eg. citrine), feldspars - 90%Quarzanteil
(u.a. Citrin), Feldspate
21.08.96/2 142 56.3 % 10.6% 33.1 % all transition forms exist; slightly fluvial reworking of the
glacially crushed/ freshly weathered material; small portion of quartz -alle Übergangsformen vorhanden; leichte fluviatile Überarbeitung des glazigen gebrochenen!frisch verwitterten Materials; geringer Quarzanteil
23.08.96/1 210 52.1 % 38.0% 9.9 % c 80 %quartz -ca. 80% Quarz
23.08.96/2 180 45.5 % 37.8 % 16.7 % heterogeneous sampie, partly important degree of rounding;
transition from glacially crushed/ freshly weathered into fluvially rounded; varieties of quartz are predominant (citrine, milky quartz) - heterogene Probe, teilweise hoher Zurundungsgrad;
Übergang von glarigen gebrochen!frisch verwittert ru fluvial gerundet; Varietäten des Quarz vorherrschend (Citrin, Milchquarz)
25.08.96/1 140 85.7 % 14.3 % - very sharp crests/ fresh fracture surfaces, freshly-edged material -
sehr scharfe Grate! iunge Bruchflächen, Material kantis-friscb
27.08.96/1 50 10.0% 80.0% 10.0 % difficult analysis, since nearly no quartz does exist, much
muscovite (mica) and brown-red aggregates -schwierige Analyse, da kaum Quarz vorhanden; viel Muskovit (Glimmer) und braun-rote ARRreRate
28.08.96/ 1 100 25.0% 55.0% 20.0% slight polishing of the fresh fracture surfaces -leichte Überpolitur
der frischen Bruchflächen
29.08.96/1 33 15.0% 85.0% - sampie with a very small portion of quartz -Probe mit sehr
geringem Quarzanteil
30.08.96/ 1 155 53.0% 20.7 % 26.3 % heterogeneous sampie, last way of transport but clearly
pronounced -heterogene Probe, letzte Transportart jedoch deutlichausgetiragt
Tab. 1:Morphometric quartz grain analysis of 10 representative sampies from south, central and west Tibet (cf. Fig. 2, 4, 6-9). Laboratory analysis (microscopy): O.A. Bauer9110/97;sampling: M. Kuhle.
Tab.1:Morphoskopische Quarzkorn-Analyse von 10 repräsentativen Proben aus Süd-, Zentral- und Westtibet (vgl. Abb. 2, 4,6-9). Laboranalyse (Mikroskopie): O.A. Bauer 9/10/97, Probenentnahme: M. Kuhle.
2, No. 13). Their ramparts enelose two glacier tongue basins.
They belonged to a west-exposed high valley glacier which has now completely melted. It came down from a massif southwest of Chomolung Kang, presently unglaciated. The end moraines are considered as being from the late-glacial up to neoglacial stages IV to V because of the ELA-depressions of at least 300 mto maximally 700mcompared with the present snow line, calculated for their ice margins. This indicates an age of ca. 5500 to 13500 yr BP (as for the chronological clas- sification of the ELA-depressions see KUHLE 1997, Tab. 1, stage IV-V). Lateral and end moraines also occur six kilome- tres to the narth, at ajunction with another parallel tributary valley at an altitude of 4100 and 4280ma.s.l. (Fig. 2, No. 14), suggesting an age of c 14250 to 13000 yr BP (stagesIIIto IV) (ibid.). 11 km up the main valley (Bo Chu), in a third orogra- phie left-hand tributary valley, two further end moraine lobes from the late-glacial to the Holocene (neoglacial) periods have been mapped (Fig. 2, IV-V below No. 19). The upper catchment area ofthis west-exposed valley belongs to Chomo- lung Kang (7312 m) and, because of its considerable altitude, it is still glaciated.
From the settlement of Kum Thang (Fig. 2, No. 11) 10 km up the Bo Chu (up to No. 15), a glacial valley extends, which is typical of south Tibet. A late-glacial to Holocene glacier- snout-gravel-floor has been filled in to form the valley bottom.
This valley bottom consisting of outwash debris is divided into gravel terraces with 3 to 5 terrace steps, each only a few metres high (No. 15). Some 25 m higher, ground moraine terraces run along both flanks of the valley (No. 11. see above). On the orographie right-hand flank of this broad trough valley, which because of its post-High Glacial (post- Last Ice Age or post-LGM) gravel floor has a box-shaped profile, glacial rock polishings have been preserved (Fig. 2, No. 16) in the form of roundings and smoothings. However, an upper ice scour limit which enables the maximum thickness of the Last Ice Age Bo Chu glacier to be recognized has not been preserved. According to our measurements in summer and autumn 1984 (KUHLE & JACOBSEN 1988), here at 4000-4400m a.s.l. freezing and thawing occurs more than 200 times per year. The daily fluctuation in the rock surface temperatures is 30 to 50°e. Under these thermal conditions an intensive post- glacial weathering of the rock must have resulted. The glacial polishings could only remain in positions long protected by lodgement till covering.
Again, the valley slopes are incised by short side valleys, so thathere, too, the main valley flank consists of the mountain spurs lying in between, which have been truncated by the main glacier. This unambiguous sequence of farms, which cannot be mistaken for convergence phenomena, provides evidence of aparent glacier many hundreds of metres in thickness, marked by important flank abrasions. The connected side glaciers flowed down from high depressions01'flat cirques in an east- exposure (No. 17). These days, outwash and mudflow fans emerge from them (No. 17). The fans are much too small to correspond to the excavation volume, i.e. to have accumulated in the course of the entire Pleistocene withouta break in accu- mulation due to any evacuating glaciation. They include dislo- cated ground mo raine material. Judging from the angle of repose, the ground moraine material may have originated to a certain extent also from the main valley flank. Corresponding post-glacial fans consisting of displaced lodgement till depo-
sited on top of the ground maraine of the main valley and interlocked with glaciofluvial gravel can also be found on the left-hand side of the valley (Fig. 2 between No. 15 and 18).
Their key position as a direct indication of former glacial land- fonns is explained for High Asia in detail by ITURRIZAGA (1999). Generally, a former glaciation is the most important factor for the development of such debris fans. It provides the loose material in the form of ground and lateral moraines on high slope positions. The side valleys contribute to displace- ment of material forming fan shapes due to their concentrated water flow. Accordingly, the most northerly of these fans contains ascattering of erratic gneiss and granite blocks (Fig.
2,No. 18).
Approximately 3 km away from this fan up the Bo Chu, there isacovering of ground moraine, metres to decametres thick. It stretches from the talweg 300 m up the orographically left slope. Its surface has almost horizontal, parallel exaration rills (28°19'N, 86°04'E at 4100 m a.s.l.). On the glacigenic round- ings of the opposite (right-hand) valley flank there are remnants of ground moraine up to just as high a level. Above these, between 4800 and 5000 m, there are high depressions in east-exposures. From their small vertical distance to the modern orographie snow line at 5600-5700 !TI, it can be concluded that they contained small glaciers or firn shields even during the late Iate-glacial period (stage IV according to KUHLE 1997, Tab. 1). Their meltwater drainage caused sander- like outwash fans and debris cones to occur on the already ice- free Bo Chu valley floor.
The two left-hand side valleys (Fig. 2, No. 21 and below No.
18) between which an Ice Age transfluence pass (No. 19) is interposed and which, therefore, during the LGM had a very thick glacier, are trough valleys with 01'without gravel floor, The narrower and steeper valley (below No. 18) has an almost elassically U'-shaped cross-profile with no flat bottom, which during the post-glacial period (Holocene) has been filled with gravel. The valley with the box-shaped trough cross-profile (No. 21) was already proved to have been formerly glaciated and shaped by the glacier ice in 1984 (cf. KUHLE 1988: 488, Fig.48).
Halfway between the junctions ofthese two left-hand tributary valleys, a large orographie right-hand side valley joins the Bo Chu (Fig. 2, No. 20). It leads down from the east side of the Shisha Pangma massif which is still highly glaciated. At the junction with this side valley, the main valley bottom is covered by ground moraine at 4120 ma.s.l.. Its thickness of probably several decametres can be concluded from the great width of the valley floor filled with moraine. Here it is more than 300 m thick. The ground moraine contains erratic boul- ders of augen-gneiss about one metre in size. The parent rock of these erratics crops out 15-20 km away on Shisha Pangma (for information on augen-gneiss petrography see KUHLE 1988: 483 and Fig. 43). Thus, at least the surface of the sedi- ment ariginated from a local moraine which was transported here by the corresponding local side glacier from the Shisha Pangma massif. This came fromaglacial catchment area at an altitude of up to 8000 m (Fig. 2, Shisha Pangma left of No.
20). In the catchment area further up the Bo Chu main valley (to the north) such an altitude is not nearly reached. Therefore, the glacier tongue of this side glacier must still have reached the valley bottom of the Bo Chu, when the nartherly Bo Chu
main glacier coming from Tibet (from Fig. 2, No. 25) - if it ever existed - had already melted. If it existeel, then most probably during the LGM; perhaps even just at the time of the earliest Late-glacial stage (stage I according to KUHLE 1997, Tab. 1), i.e., when the ELA had decreased by somewhat more than 1000 m compared with the present-day level. For this reason, the local ground moraine we are talking about must be geomorphologically dated as a milddie or 1ate late-glacial deposit (Fig. 2, No. 20, II-IV = stages II-IV, c 15000-13000 yr BP). Should this northerly main valley glacier have existed, its ground moraine wou1d be underneath this young cover of ground moraine. Otherwise, in the case of a non-existent main valley glacier from Tibet, down from the 5060 m high Yagru Xiong La (No. 25), lee Age alluvions (grave1 floors) wou1d have to lie beneath it.
Some 2.5 km further up the Bo Chu, an orographie left-hand side glacier (No. 21) from the 7312 m high Chomolung Kang, might, in an analogous fashion, have reached the Shisha Pangma east glacier even later than the Bo Chu parent glacier.
The methodologically most important information in this chapter is as folIows: Just like a large side valley glacier from the Fuqu Chu coming down from the south flank of Shisha Pangma could have built up the lower Bo Chu glacier, the side glacier from the east of Shisha Pangma could have built up the middle Bo Chu glacier. In both cases, the high catchment areas would be the strongest argument in favour of such a glacial feeding. However, the precipitation from the windward side in the south towards the Shisha Pangma east slope is already noticeably decreasing. On the other hand, on the east slopes, the lower amount of incoming radiation favours formation of the glacier. Accordingly, the main question that now must be answered is whether the northerly upper Bo Chu was ever glaciated. Its glaciation could only have taken place from Tibet.
Indications of a past glaciation of the upper Bo Chu and its connection to the plateau ice from Tibet
Up-valley from the junction of the side valley from Shisha Pangma (Fig. 2, No. 20), the Bo Chu has a classic glacigenic- ally-shaped box cross-profile (No. 22) over a distance of more than 20 km. It has been developed from a broad glacial U-shaped valley because of the sedimentation of loose mate- rial like ground moraine anel, after deglaciation, gravels at its bottom. Nine kilometres up-valley from the junction of the left-hand side valley which comes down from Chomolung Kang (No. 21), a ground mo raine cover metres to decametres thick was mapped on the orographie right-hand slope (28°
28'N, 86°09'50"E, 4310 m a.s.l., Fig. 2, No. 23). It lies on top of smoothly polished outcropping sedimentary rocks, whose surfaces show a pattern of parallel, horizontal striations. This is characteristic of a valley glacier which due to its movement and the boulders ofthe subglacial mo raine frozen to its bottom and edges ploughs through its own, partly already consoli- dateel, ground moraine. At the slope foot the moraine mantle was undercut by the lateral erosion ofthe Sun Kosi river. Here, eroded rills and earth pyramids, typically associated with them, developed. These were formed from more consistent material between the rills, i.e. residually. The material which can, therefore, only be approached as ground moraine is also preserved on the orographic left-hand slope. It reaches on both
sides of the valley to c 400 m above the present valley floor.
This proves a minimum altitude of the glacier trimline of c 4700 m a.s.1.. Glacigenic flank polishings, however, come up to the culminations ofthe valley flanks.
The valley cross-profile under discussion is situated at the true head of the Bo Chu valley. Here, the two source branches of this main valley meet at an obtuse angle of 105°. The orogra- phie left-hand branch is still called the Bo Chu. The right-hand one is the Yagru Chu. In the triangle they form is the Yagru Xiong La (also Sho La or Lalung La; Fig. 2, No. 25). Up there at 5060 m, the steep slope on the south edge of the Tibetan plateau begins.
What is the significance of a glacier thickness weil over 400 m (see above), here at the root ofthe Bo Chu (No. 23)7 The ice must have crossed this tri angular inset between the two SOUlTe valley branches directly from the Tibetan plateau. In the Bo Chu main valley which begins here, it then collected to form a south Tibetan outlet glacier. This ice supply from Tibet derived from the ground moraine at the main valley head can be empi- rically verified all over this triangle of plateau. Here, a ground moraine with erratic granite boulders is also to be found (No.
24). In some places, it has become exposed in young fluvial rills. Sedimentary bedrock is in the underground. In other places, as in sub ordinate talwegs, the moraine is covered by gravel interspersed with pebbles and sand. These have been washed out from the moraine surfaces of the local catchment areas of those rills and small valleys during the deglaciation and post glacial periods. The ground moraine spreads over the entire surface to far above 5000 m. It uniformly covers the Yagru Xiong La (5060 m No. 25) and also the neighbouring kilornetres-wide highland areas of south Tibet (No. 26). The polymictic, mostly erratic boulders are roundeel, facetted or have rounded edges. They "float"isolated from each other in a matrix of fine material. This is also clearly morainal due to the bimodal grain size distribution (Fig. 6). The c 9%clay of the fine grain peak characterizes the moraine as ground moraine, as does the predominance of 56.3 % quartz grains which by morphoscopic microscopic analysis is shown to be "glacially crushed" (Fig. 5, 21.08.96/2). The alternative of convergent material which would also be interpreted as "freshly weath- ered" can be rejected. The material at the culmination of the pass (No. 25) must, because of the horizontal topography and its erratic composition, be far-travelled substrate. The lack of any dip in the slope hinders even any incorporation of in situ weathered material from the out-cropping rock. Two or three kilometres beyond this culmination, glacially streamlined hills have developed in the sedimentary rock on a basal area at 4800 m a.s.l. (Fig. 2, obliquely to the left above No. 26). They are also covered by ground moraine with erratic granite boul- ders (No. 26). Further to the north there are roches moutonnees and large streamlined glacigenic erosive forms also in the granite bedrock. These are the source areas for (i) the surrounding local mo raine (No. 27), (ii) ground moraine with erratics (No. 26) transported in a southerly direction upwards towards the pass (No. 25) and (iii) isolated large erratic granite boulders without a moraine mantle (above No.
26) but also (iv) far-moraine transported over the Yagru Xiong La into the Bo Chu (No. 23) (see above). The sedimentary loose rocks and also the erosive forms described provide evidence of a complete covering of this region of south Tibet by a former glacier.
CUMULATIVE FREQUENCY GRAIN-SIZE CUAVE 21.08.199612
< 2 2 - 6 6 - 2020 - 60 60 - 200 - 600- 200 600 2000 (DIAMETER 1/1.000)
HUMUS CONTENT: 3,73 % L1ME CONTENT: 0,32%
Fig. 6: Ground moraine matrix taken from a depth of 0.15 m at 5060 m a.s.1.
in the high plateau area of the Yagru Xiong La (cf. Tab, I, Fig. 5: 21.08.96/2).
Locality: Fig. 2, NO.25. The characteristic bimodal course of the curve shows a fine grain peak in the clay fraction, typical of ground moraines. Polymiet er- ratic boulders of granite, quartzite and gneiss arc incorporated into the humus- containing ground moraine which is slightly weathered on the surface. Therc are also several limestone components. Metamorphie bedrock occurs in the underground.
Abb.6:In 5060m ü.M. im Bereich der Hochplateaufläche des Yagru Xiong La aus 0,15 mTiefe entnommene Grundmoränenmatrix (s. Tab, I, Fig. 5:
21.08.96/2): Lokalität: Fig. 2, Nr. 25. Der charakteristische bimodale Kurven- verlaufmit einem zweiten Maximum in der Tonfraktion ist typisch für Grund- moräne. Die oberflächlich leicht verwitterte, hurnus-haltige Grundmoräne enthält erratische polymikte Blöcke aus Granit, Quarzit und Gneis; auch eini- ge Kalkkomponenten sind vorhanden. Im Untergrund stehen Metamorphite an,
Before discussing this result, further field data from south Tibet are to be reported. We are now 15-18 km north of the local watershed between the upper Bo Chu and a few shallow basin-like side valleys which run to the north towards Xaga Chu and into Tibet (No. 28). The gentle slopes (8-15°) down to Xaga Chu are completely covered by ground moraine between c one and several metres thick (Fig. 11) which, typically, consists for the most part of fine matrix. This contains, iso- lated from each other, polymictic boulders which are fist- sized up to at most head-sized, roundeel, facetted or with rounded edges. Among them are also granite erratics. In the underground is sedimentary bedrock. The surface of this lodgement till covering is conspicuously shapeless and smooth (Fig. 11 from • to • in the back ground). Itcovers these gentle slopes right up to the towering hills over 5000 m high, south east ofXaga Chu, and even sometimes covers their culminations (Fig. 11 .. ). This provides evidence of their complete former ice covering which can also be diagnosed from their rounded form. From this, the minimum height of the Ice Age glacier surface can be deduced (Fig. 11---). In situ scree slopes which would have been characteristic of a perigla- cial environment continuing over many hundreds of thousands of years is missing on the slopes (Fig. 11.on the left). This gives a further indirect indication ofIceAge glaciation. Lastly, the 2-4 m deep rills must be mentioned. They are cut into the lodgement till by the present-day run-off of rain water (e.g. at 28°38'-45'30"N, 86°06'-1 O'30"E). These new fluvial forms due to the modern climate destroy the former cover of loose rock.
This is also proof of a completely different type of morphody- namics during the Ice Age.
A ground moraine exposure deriving from backward erosion ofrills (Fig. 2, No. 28; Fig. 11) lies 625 m lower than the water divide on the Yagru Xiong La (No. 25). An only local, small-
scale glacier cover would not, on its own, render the thickness of ice (see above) deduced from the shapes of the hills and their moraine covers possible, taking into account such a great difference in altitude. Accordingly, an ice thickness of more than 625 m is necessary. The wide, extensively levelled ground moraine cover which lies over and around the relief ofthe hills also speaks in favour of this.
A further indicator is the glacigenic trough shape of the Xaga Chu (Fig. 2, No. 29). Its bottom is covered by a layer of gravel made up during the late late-glacial, the neoglacial period and in historical times. The Xaga Chu drains the north side of the Shisha Pangma massif and, therefore, the glaciofluvial gravel floor which is over one kilometre wide and has been develop- ing since deglaciation, can be correlated with the late late- glacial to historical glacier stages IV to c IX (13500-180 yr BP according to KUHLE 1997, Tab. 1) in the Shisha Pangma north- ern flank (for the glacial history of the Shisha Pangma north- ern slopes see KUHLE 1988: 468 Tab. 1 and 479-487). Thus, not only the erosive trough shape becomes understandable, but also the later accumulation of gravel due to the more recent glacier history since the LGM. There is no other alternative.
On the trough flanks of the Xaga Chu, the ground mo raine described above continues for decakilometres down-valley towards the north-east (Fig. 2, No. 30). As the slopes become steeper, the rills dissecting them become closer. Obviously, these fluvial channels, still shallow for the time being, only started to interglacially reshape the large-scale abrading and also smoothing glacial geomorphology a few thousand years ago. Fluvial geomorphodynamics throughout the whole of the Pleistocene would have left behind a completely different, namely a V-shaped valley landscape divided up into smaller areas.
The glaciogeomorphological results from the southern edge of the Tibetan high plateau and the southerly adjoining upper Bo Chu lead to the following picture. On top of the plateau there was an ice which completely covered the hilly relief. Its thickness was so great that it even filled large valleys north of the water divide, like the Xaga Chu, completely. Large-scale ground moraine covering and the distribution of granite erra- tics from north ofthe water divide over the Yagru Xiong La in a southerly direction down into the Bo Chu are evidence of a former ice flow from Tibet. Thus, such a Bo Chu outlet glacier (Fig. 1, right-hand side, or east of Shisha Pangma) arose from a south Tibetan upland ice (Fig.l,13; cf. also Fig. 3, near the right-hand edge: Tibet Himalaya= south Tibet). It was chan- nelled by the Bo Chu and drained off through the Himalayan breakthrough valley. In the Himalayas, the Chomolung Kang, the Shisha Pangma and the Rolwaling Himal (below Fig. 2) fed this glacier. The glaciogeomorphology of the Man-k'o-pa basin which joins the Xaga Chu in the northeast (Fig. 2, ob- liquely to the right above No. 30) affects the evidence of a Bo Chu outlet glacier only indirectly. This was described in a previous study (KUHLE 1988, Tab. 1, Fig. 1 and 31).
EVIDENCE FOR A PROBABLE LAST lCE AGE (LGM) KYETRAK OUTLET GLACIER
A further glaciogeomorphological key locality is the Kyetrak Chu NNW ofthe Cho Oyu (Fig. 2). In this context, it is not the
CUMULATIVE FREQUENCY GRAIN-SIZE CURVE 23.08.1996/2 CUMULATIVE FREQUENCY GRAIN-SIZE CURVE 23.08.1996/1
HUMUS cONTENT: 1,89 % L1ME CONTE NT: 6,99%
Silt Sand
--+---,==-.-j 100
f---+c=~-:---..---___t80
o - - - j60 %
o--,~~-_____t40
20
o
Clay50f - - t - - - - 40
"k 30j - - - j - - - -
°20 j----.-+--~---c.-,z'c+.
10h~+-
o
L1Jjtr:r=:::i=~~Fig. 7: At 4725 m a.s.l., late glacial to historie ground moraine taken from a depth of 0.1 m on the bottom of the upper Kyetrak valley left of the gravel floor, 2 km away from the tongue of the Kyetrak glacier. Locality: Fig. 2, No 32. The minor amount of the clay portion is to be reduced to the important portions of localmoraine which has bcen transported not too far. These large portions are typical of the narrow valley landscape of the Himalaya, where the sample locality is situated. See Tab. I, Fig. 5,23.08.96/1.
ICIe.y
i
Silli
Send501---:---~:- 100
%;1 ff~:,< 2 i 11i 'I~~~ %
2 - 6 6 - 2020 - 60 60 - 200 - 600- 200 600 2000 (DIAMETER 1/1.000)
Abb. 7: In 4725 m Ü.M. im Talboden links der Schottersohle spätglaziale bis historische Grundmoräne im oberen Kyetrak-Tal, 2 km von der Zunge des Ky- etrak-Gletschers entfernt, aus 0, I m Tiefe entnommen. Lokalität: Fig. 2 NI'.
32. Die geringen Ton-Anteile werden auf die bedeutenden Anteile nur wenig transportierter Lokalmoräne zurückgeführt. Diese ihrerseits sind für die enge Tallandschaft des Himalaya, in der sich die Probenlokalität befindet, typisch.
S. Tab. I, Fig. 5,23.08.96/1.
<2 2 - 6 6 - 2020 - 60 60 - 200 - 600- 200 600 2000 (DIAMETER 1/1.000)
HUMUS CONTE NT: 2,58 % L1ME cONTENT: 0,15% A continuation of the glacier reconstruction south of the
Nangpa La is not necessary in this context. Much more impor- tant is the question of where the ice came from. Did it flow down exclusively from Cho Oyu, Gyachung Kang and from the neighbouring peaks on the southern side of the Himalayas or did it also come over the water divide from Tibet? In this connection it becomes significant that the Kyetrak valley axis is a direct outlet from south Tibet. At present, it leads down from Nangpa La in a northerly direction (see above), i.e. into the upland (Fig. 2, from No. 31 to 39 and further on towards Ting-Jih), Thus, the problem is the following: Was there a former outlet glacier which flowed over the water divide counter to the slope of the valley? Or,put in another way: Did a thick Tibetan inland ice flow over the bordering Himalayan passes?
Traces of the maximum former thickness of ice in Kyetrak Chu present-day direction of descent of the valley which is of importance but rather the direction of its axis. This runs from north to south right through the Himalayas from Tibet. The Kyetrak valley ascends first from the settlement ofTing-Jih to the recent glacier pass Nangpa La (or Khumbu La; Fig. 2, from No. 39 to 31) at an altitude of 5717 m. Beyond the Hirna- layan water divide (Fig. 2, south of No. 31) the axis continues along the southern Himalayan slopes. There, on the other side ofthe pass, the valley ofNangpo Dzango and further down the Dudh Kosi valley, run down through the Khumbu Himal. The reconstruction of its maximum Ice Age glaciation has been taking place since 1956.Ithas led to the consistent result that the lowest end of the glacier must have reached down at least to 1580 m a.s.l. and, therefore, into the valley chamber of Surke in the Dudh Kosi (HEUBERGER 1986: 30; KUHLE 1985, 1987). Where the two tributaries, the Nangpo Dzango glacier from Nangpa La (No. 31) and the Imja Drangka glacier flowed together to form the Dudh Kosi parent glacier (Khumbu parent glacier) (27°53'N, 86°44'E), this had a thickness of at least 600-850 m (KUHLE 1987: 406, Fig. 19:
407/408).
The present-day Kyetrak glacier, 10 km long, flows west ofthe Cho Oyu (Fig. 12, No. 1) from the 5717 m high Nangpa La (Fig. 2, Fig. 12, No. 31) in a northerly direction along the bottom of the Kyetrak valley (Fig. 12) down to 4800m a.s.l.
(Fig. 2, below No. 32). 1nvestigations of the orographie left- hand, west valley flank have shown that the sedimentary bedrocks (Fig. 120) are covered with ground moraine. At 5250m (Fig. 2, No. 33), approximately 600 m above the valley floor (Fig. 12, - 5 on the left), a sampie of the matrix of the ground moraine was taken (Fig. 9). The bimodal grain-size distribution and the fine grain peak in the clay
Cl
0 %) prove their morainic character. The exposed position of this material on a mountain ridge ren der an alternative diagnosis of fluvial accumulation impossible. The morphoscopic grain analysis (Tab. 1, Fig. 5,25.08.9611) yielded a predominance of 85.7 % glacially crushed/fresh1y weathered quartz grains typical of ground moraine. They cannot have been generated by fresh weathering in situ for the following reasons: The sampie was taken from an earth pyramid which was several metres above the bedrock of the underground (Fig. 12 ... ). The form"earth pyramid" is proof of far-travelled material. The remains of earth pyramids themselves, as have been studied in the
Fig. 8: Sampling at 4730 m a.s.l. from the orographie left-hand late glacial ground moraine-01'lateral moraine terrace (Stage III or IV?) c 90 m above the recent gravel floor of the Kyetrak valley. Depth: 0.15 m; locality: 28°22'N.
86°37'50"E. Fig. 2. No. 40. The bimodal course ofthe cumulative curve, char- acteristic of moraines, is obvious. The important differences of the carbonate content (er. Fig. 7, 9) within the moraine sediments of one and the same valley immediately show the strong topographie dependence of the ice flow on the thicknesses ofthe glacier. See Tab. I, Fig. 5,23.08.96/2.
Abb, 8: In 4730 m ii.M. aus der orographisch linken spät-eiszeitlichen Grund- moränen- oder Ufermoränenterrasse (Stadium III od.Iv") ca. 90 m Über der rezenten Schottersohle des Kyetrak-Tales entnommen. Entnahmetiefe: 0,15 m;
Lokalität: 28°22'N, 86°37'50"E, Fig. 2 Nr. 40. DerfürMoränen kennzeichnen- de bimodale Summenkurvenverlauf ist deutlich. Die bedeutenden Unterschie- de im Kalkgehalt (vgL Fig. 7, 9) innerhalb von Moränensedimenten ein und desselben Tales bilden die starke topographische Abhängigkeit des Eisflusses von den Gletschermächtigkeiten unmittelbar ab. S. Tab. 1, Fig. 5, 23.08.96/2.
European Alps (e.g. on the Ritten or near Meran at Tirol Castle) and in the Karakorum in classic moraine localities, are evidence of ground moraine as initial material. A proportion of 85.7 %(Fig. 5, 25.08.9611) of matrix with grains that have been glacially crushed is an extremely high value also for Tibetan ground moraine. The reason for this is the very high relief position which has prevented post-genetic reworking almost comp1etely (cf. below).
CUMULATIVE FREQUENCY GRAIN-SIZE CURVE 25.08.1996/1
jClay
I
Sllt Sand5 0 8 '- - - - 100
40 - . - - - 1 8 0
0;. 30 - - - -
----i
60 %o
~~ ~
.•: -···..-=--1.0rl i~
<2 2 - 6 6 - 20 20 - 60 60 - 200 - 600- 200 600 2000 (DIAMETER 1/1.000)
HUMUS CONTENT: 3,79 % L1ME CONTENT: 17,32 %
Fig. 9: At 5250 m a.s.l. sampling of high- to late glacial ground moraine at a depth of 0.1 mon the orographie right-hand flank of the Kyetrak valley, c 600 m above the present talweg. Locality: Fig. 12... : Fig. 13'" right hand, back- ground; Fig. 2 No. 33. Thebirnodal course ofthe cumulative curve with a pro- nounced fine grain peak in the clay is typical ofthe glacigenic character ofthe sediment. The moraine contains large erratic granite boulders; it covers the sand- and schist bedrocks extensively. See Tab. I, Fig. 5,25.08. 96/l.
Abb, 9: Aus 5250 m Ü.M. in der orographisch linken Flanke des Kyetrak-Tales ca. 600 m Über der heutigen Tiefenlinie aus hoch- bis späteiszeitlicher Grund- moräne aus 0,1 m Tiefe entnommen. Lokalität: Fig. 12... , Fig. 13'" rechts hinten; Fig. 2 Nr. 33. Der bimodale Verlauf der Summenkurve mit dem ausge- prägten Feinkornpeak im Ton ist für den glazigenen Charakter des Sediments kennzeichnend. Die Moräne enthält große erratische Granitblöeke und deckt großflächig anstehende Sand- und Schiefergesteine ab. S. Tab. I, Fig. 5, 25.08.96/1.
A further proof is obtained with the help of erratic granite boulders in the moraine 01' on the sedimentary bedrock without any surrounding fine material matrix (Fig. 12 0, ':::.I ':::.I). These granite boulders and ground moraines (Fig. 12 ':::.I, ... ; Fig. 13 0; 1) were mapped on the west flank of the Kyetrak Chu over distances of kilometres on slopes, valley shoulders and mountain ridges, covering the reddish silt- and sandstone bedrocks (Fig. 12 0,Fig. 13) as wellas the light- coloured limestone rocks (Fig. 2, on the right and diagonally right above No. 33, between No. 34 and 40). Their highest occurrence lies at least 700 m above the level of the valley floor. Thus, the upper level ofthe past Kyetrak glacier which is indicated by glacigenic accumulations, lay at an altitude of 5500 m. A glacier cannot, however, accumulate moraine and bou1ders above the snow line. There, glacia1 erosion and debris transport predominate. At an altitude of 5500m,we are only 500 m below the present-day snow line (ELA) of the Kyetrak glacier. During the LGM, the snow line was 1000-1200 m lower than it is now, so that these highest accumulations were at least 500 m above the lee Age snow line (ELA). Therefore, these moraines and erratic granite boulders must have a1ready been a Iate-glacial accumulation. This also means that the maxirnum Last lee Age (LGM) thickness of the glacier cover must have been much greater. Its surface lay much higher than 5500 m, i.e. probably at 5700-5900 m. Geomorphologica1 indications for this conclusion are the mountain ridges and
Fig.11: View from 4435 m a.s.l. (Fig. 2, No. 28; 28°38'N, 86°06'E) into the orographie right-hand flank ot the Xaga Chu facing east, The exposure shows a gro- und moraine cover (. foreground) with coarse polymiet boulders floating in a matrix wich contains great portions of clay. The boulders(0)consist of granite and quartzite. The granite is erratic, beeause sedimentary bedrock is in the underground(0).The boulders are rounded at the edges, partly glacigenically facetted (0).The ground mo raine sheet (.), part ofwhich is far-travelled, mantles the flat slopes (from the foreground up to the background on the right.) as well as the steeper foot slopes (. background on the left) ofthe mountain ridges rounded by the glacier ground scouring ("'). (----) marks the minimum height ofthe Ice Age inland ice surface, deduced from the field data. Photo M. Kuhle.
Abb. 11: Auf 4435 m Ü.M. (Fig. 2, Nr. 28; 28°38'N, 86°06'E) in die orographisch rechte Flanke des Xaga Chu nach Osten fotografiert. Der Aufschluss zeigt eine Grundmoränendecke (. Vordergrund) mit groben, polymikten Blöcken, die in einer stark tonhaltigen Matrix "schwimmen". Es sind Granit- und Quarzitblöcke (0),wobei erstere erratisch sind, da der Untergrund aus Sedimentgestein besteht(0).Die Blöcke sind kantengerundet. teilweise auch glazigen facettiert(0).
Die zum Teil ferntransportierte Grundmoränendecke (.) bekleidet die flachen Hänge (Vordergrund bis Hintergrund rechts) sowie die steileren Fußhänge (. Hin- tergrund links) der vom Gletscher-Grundschliffabgerundeten Bergrücken ("'). (u__) markiert die Mindesthöhe der eiszeitlichen Oberfläche des Inlandeises, die von den Geländedaten abgeleitet wurde. Foto M. Kuhle.
cupolas above the west flank of the Kyetrak valley - devoid of moraine and/or erratic boulders - which are polished round up to the altitudes mentioned (Fig. 12 and 13 ... on the right; Fig.
2, No. 34 and 35).
On the opposite east flank of the Kyetrak valley, classic flank polishings have been preserved up to the same altitude (5700- 5900 m). This concerns characteristic forrns of glacigenic flank abrasion, such as rochemoutonnee-like ridges situated high up (Fig. 12 ... directly belowNoAand 8; Fig. 2, between No. 32 and 36), back-polished glacially truncated spurs with rounded edges and rock crests (Fig. 2, No. 37, left above No.
36 and on the right to diagonally right above No. 32; Fig. 12 white left) and a striped flank polishing due to exaration. Its lineation traces the outcrops of the strata. (Fig. 12 ... on the very left; Fig. 2, diagonally left below No. 36). In many places, the two forms (outcrop and polishing) are combined and interfere with each other. In the Kyetrak valley, this so- called "Schichtkopfstreifenschliff" (outcrop strip polishing) (cf. VON KLEBELSBERG 1948/49) has marked the tri angular- shaped slopes of back-polished spurs with a pattern of linea- tion towards the south (Fig. 13 ... on the left).
On1y in a few places is the upper east valley flank:, character-
ized by removal, covered with remains of moraine (Fig. 12 .. "). Their decametres-thickness is recognizable from afar due to the earth pyramids and the rills between them. Here, late-glacial ground moraine material is concerned. It was left by loca1 eastern side glaciers which joined the Kyetrak glacier north of Cho Oyu and Gyachung Kang (Fig. 12, No. 1 and 2).
The assumption that there were junctions is based on the very high positions of the moraines at the exits of the side valleys, several hundred metres above their valley floors. The upper border of the flank: polishings on the eastern slopes of the Kyetrak valley allows the diagnosis of a continuous polishing limit, falling slightly away from north to south (Fig. 12 -mO 0- ---, Fig. 13 ----0). This can only be the minimum altitude ofthe Ice Age (LGM) trimline of the Kyetrak glacier. The maximum ice level was probably only reached for a short time and is, therefore, geomorphologically hard1y discernib1e - which is why its late-glacial and post-glacial complete obliteration is probable and any remaining signs of it improbable.
The Kyetrak valley was, therefore, filled by a glacier at least 1000 m thick during the Last 1ce Age whose surface (Fig. 12 and 13:----0 0----, ----; 0) was inclined counter to the inclina- tion ofthe valley floor. This means an ice drainage direction at that time towards the south over the Himalayan water divide,
Fig. 13: Taken at 5300 m a.s.l. from the orographie left-hand flank ofthe still very wide Kyetrak Chu (valley) (Fig.2, right ofNo. 34) to the south. There are er- ratie granite boulders, partly well-rounded, in the foreground (0,sitting person for seale). They lie on surfieially weathered reddish bedroek sandstones. Angular loeal moraine boulders of limestone, moved only a little, are also preserved on sandstone. (. below ----0; I, 1I, III and IV) are ground moraines and lateral mo- raines ofthe LGM to late-glaeial.
I."
on the right is shown in detail in Figure 12. ( .. ) mark mountain ridges, round-polishedby the High Glacial glacier ice, which on the orographie left-hand valley side have been formed without exeeption in the outeropping edges ofmetamorphie sedimcntary rocks ( .. on the right).(----0) indicate the minimum height ofthe High Glaeial glaeier levels, deduced from this geomorphology. Photo M. Kuhle.
Abb. 13: Foto aus 5300 m Ü.M. von der orographisch linken Flanke des noch immer sehr breiten Kyetrak Chu (Tal) (Fig. 2 rechts von Nr. 34). Erratische Granit- blöcke, teilweise gerundet, sind im Vordergrund zu sehen(0,sitzende Person als Größcnvergleich). Sie liegen auf anstehenden, oberflächlich verwitterten, rötli- chcn Sandsteinen. Auch eckige lokale Moränenblöcke aus Kalk sind auf dem Sandstein erhalten. ( • unter ----0; I, Il, III and IV) sind Grund- und Ufermoränen vom Hochglazial (LGM) bis Spätglazial. I . . . rechts ist in Abbildung 12 detailliert abgebildet. ( .. ) markiert vom hochglazialen Gletschereis gerundete Berg- rücken, welche auf der orographisch linken Talseite ohne Ausnahme in Schichtköpfen von metamorphem Sedimentgestein ausgebildet sind ( .. rechts). (----0) bezeichnet die von dieser Geomorphologie abgeleitete Mindesthöhe der hochglazialen Gletseherstände. Foto M. Kuhle.
